Summary Sepsis is a major health issue with over 1.5 million cases of sepsis/year and >250,000 deaths/year in the US. All drugs recently developed in animal models to treat sepsis have failed in clinical trials. Thus, there is an urgent need for the development of models that better represent the human disease and new technologies to screen potential therapeutics for the treatment of sepsis to efficiently predict their response in humans. To address this need, we developed and extensively validated a novel 3D biomimetic microfluidic assay (bMFA) for characterization of the leukocyte adhesion cascade that realistically reproduces a microvascular network with accurate geometry from microvascular networks observed in vivo. We propose to use this novel bMFA as a platform to identify omic and phenotypic species differences in drug responses between mice and humans to provide important insight into the failure of some compounds tested in mice to translate into viable therapeutics in humans. We will use a combination of microfluidic, omic, and in silico models to determine how and to what degree the bMFA reproduces the inflammatory response resulting from sepsis in animal models and its specific response to relevant therapeutics. We will then study the response of human cells from normal and septic patients to pro-inflammatory mediators and therapeutics. In Aim 1, we test the hypothesis that phenotypic differences impact neutrophil-endothelial interactions similarly in vivo and in vitro. In Aim 2, we test the hypothesis that species variances differently influence neutrophil-endothelial interactions in sepsis. In Aim 3, we test the hypothesis that the response of murine cells to two candidate therapeutics (BN-52021 platelet-activating factor receptor antagonist and a novel PKC? inhibitor) in vitro is predictive of the response in an in vivo model of sepsis. Relevant functional and proteomic changes will be compared between bMFA and the mouse model. We will determine whether disease state differentially affects the response to these therapeutics in cells from septic mice as compared to cells from sepsis patients. Employing both human and mouse cells, we will use in silico modeling to examine inflammatory signaling for differences between species and to determine how different therapeutics impact the progression of inflammatory signaling in sepsis. These studies will provide a unique technology by which critical human (ex vivo) data can be used to predict whether preclinical animal models are likely to predict outcome of clinical trials. The long- term goal of this project is to develop appropriate in vitro models for basic understanding of cellular mechanisms and rapid pre-screening and efficient selection of promising therapeutics that will likely be efficacious in human trials.